Optimization and validation of the international metabolic prognostic index for CD19 CAR-T in large B-cell lymphoma

Abstract

While CD19-directed CAR T-cell therapy represents a transformative immunotherapy for relapsed/refractory large B-cell lymphoma (r/r LBCL), more than 50% of patients ultimately progress or relapse. Recently, the International Metabolic Prognostic Index (IMPI) - incorporating age, stage, and metabolic tumor volume (MTV) - was shown to improve prognostication for LBCL frontline treatment. Here, we examine its utility to predict toxicity and survival in CAR-T recipients. This multicenter observational study spanning six international sites included 504 patients with available 18FDG-PET/CT imaging at last response assessment prior to lymphodepletion. Optimal CAR-adapted MTV thresholds were identified in a development cohort (n = 256) and incorporated into a CAR-T-specific IMPI ("CAR-IMPI"). The prognostic performance of CAR-IMPI was validated in an independent cohort (n = 248). CAR-IMPI risk categories, defined by the median (1.35) and terciles (1.07, 1.58), demonstrated significant discrimination for progression-free survival (PFS; p < 0.0001) and overall survival (OS; p < 0.0001) in both cohorts. Multivariate Cox regression confirmed CAR-IMPI as an independent predictor of survival, accounting for pre-lymphodepletion LDH and CRP, performance status, treatment center, and CAR-T product. Patients in the CAR-IMPI high-risk category experienced increased severity of CRS and ICANS, and higher rates of intensive care unit (ICU) admissions. In an exploratory analysis, combining CAR-IMPI with established indices of high-risk systemic inflammation (CAR-HEMATOTOX, InflaMix) further enhanced survival stratification. The CAR-IMPI may provide a potent and validated PET-based tool for risk stratification of clinical outcomes in patients with r/r LBCL receiving CD19 CAR-T therapy. Our data highlight the utility of combining clinical and radiological modalities, with implications for patient selection and the anticipated level-of-care for toxicity management.